[llvm] 3538ee7 - [CostModel][X86] Improve AVX1/AVX2 v16i32->v16i16/v16i8 truncation costs (PR51972)
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Sun Sep 26 05:51:06 PDT 2021
Author: Simon Pilgrim
Date: 2021-09-26T13:43:46+01:00
New Revision: 3538ee763d13a26515a224ddeb3a51a5af143e38
URL: https://github.com/llvm/llvm-project/commit/3538ee763d13a26515a224ddeb3a51a5af143e38
DIFF: https://github.com/llvm/llvm-project/commit/3538ee763d13a26515a224ddeb3a51a5af143e38.diff
LOG: [CostModel][X86] Improve AVX1/AVX2 v16i32->v16i16/v16i8 truncation costs (PR51972)
Based off worst case btver2 (AVX1) and haswell (AVX2) llvm-mca reports
Added:
Modified:
llvm/lib/Target/X86/X86TargetTransformInfo.cpp
llvm/test/Analysis/CostModel/X86/arith-fix.ll
llvm/test/Analysis/CostModel/X86/arith-overflow.ll
llvm/test/Analysis/CostModel/X86/cast.ll
llvm/test/Analysis/CostModel/X86/min-legal-vector-width.ll
llvm/test/Analysis/CostModel/X86/trunc.ll
Removed:
################################################################################
diff --git a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
index 52e8238771eb6..8b40df5dff3df 100644
--- a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
+++ b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
@@ -1940,6 +1940,8 @@ InstructionCost X86TTIImpl::getCastInstrCost(unsigned Opcode, Type *Dst,
{ ISD::TRUNCATE, MVT::v8i1, MVT::v8i32, 2 },
+ { ISD::TRUNCATE, MVT::v16i16, MVT::v16i32, 4 },
+ { ISD::TRUNCATE, MVT::v16i8, MVT::v16i32, 4 },
{ ISD::TRUNCATE, MVT::v16i8, MVT::v8i16, 1 },
{ ISD::TRUNCATE, MVT::v16i8, MVT::v4i32, 1 },
{ ISD::TRUNCATE, MVT::v16i8, MVT::v2i64, 1 },
@@ -2015,6 +2017,8 @@ InstructionCost X86TTIImpl::getCastInstrCost(unsigned Opcode, Type *Dst,
{ ISD::TRUNCATE, MVT::v8i1, MVT::v8i64, 9 },
{ ISD::TRUNCATE, MVT::v16i1, MVT::v16i64, 11 },
+ { ISD::TRUNCATE, MVT::v16i16, MVT::v16i32, 6 },
+ { ISD::TRUNCATE, MVT::v16i8, MVT::v16i32, 6 },
{ ISD::TRUNCATE, MVT::v16i8, MVT::v16i16, 2 }, // and+extract+packuswb
{ ISD::TRUNCATE, MVT::v16i8, MVT::v8i32, 5 },
{ ISD::TRUNCATE, MVT::v8i16, MVT::v8i32, 5 },
diff --git a/llvm/test/Analysis/CostModel/X86/arith-fix.ll b/llvm/test/Analysis/CostModel/X86/arith-fix.ll
index fcb2b90de81ca..f9aaa427751ac 100644
--- a/llvm/test/Analysis/CostModel/X86/arith-fix.ll
+++ b/llvm/test/Analysis/CostModel/X86/arith-fix.ll
@@ -81,8 +81,8 @@ define i32 @smul(i32 %arg) {
; AVX1-NEXT: Cost Model: Found an estimated cost of 114 for instruction: %V16I32 = call <16 x i32> @llvm.smul.fix.v16i32(<16 x i32> undef, <16 x i32> undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I16 = call i16 @llvm.smul.fix.i16(i16 undef, i16 undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call <8 x i16> @llvm.smul.fix.v8i16(<8 x i16> undef, <8 x i16> undef, i32 3)
-; AVX1-NEXT: Cost Model: Found an estimated cost of 53 for instruction: %V16I16 = call <16 x i16> @llvm.smul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
-; AVX1-NEXT: Cost Model: Found an estimated cost of 106 for instruction: %V32I16 = call <32 x i16> @llvm.smul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
+; AVX1-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V16I16 = call <16 x i16> @llvm.smul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
+; AVX1-NEXT: Cost Model: Found an estimated cost of 90 for instruction: %V32I16 = call <32 x i16> @llvm.smul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I8 = call i8 @llvm.smul.fix.i8(i8 undef, i8 undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V16I8 = call <16 x i8> @llvm.smul.fix.v16i8(<16 x i8> undef, <16 x i8> undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V32I8 = call <32 x i8> @llvm.smul.fix.v32i8(<32 x i8> undef, <32 x i8> undef, i32 3)
@@ -100,8 +100,8 @@ define i32 @smul(i32 %arg) {
; AVX2-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V16I32 = call <16 x i32> @llvm.smul.fix.v16i32(<16 x i32> undef, <16 x i32> undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I16 = call i16 @llvm.smul.fix.i16(i16 undef, i16 undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V8I16 = call <8 x i16> @llvm.smul.fix.v8i16(<8 x i16> undef, <8 x i16> undef, i32 3)
-; AVX2-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %V16I16 = call <16 x i16> @llvm.smul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
-; AVX2-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V32I16 = call <32 x i16> @llvm.smul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
+; AVX2-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %V16I16 = call <16 x i16> @llvm.smul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
+; AVX2-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %V32I16 = call <32 x i16> @llvm.smul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I8 = call i8 @llvm.smul.fix.i8(i8 undef, i8 undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V16I8 = call <16 x i8> @llvm.smul.fix.v16i8(<16 x i8> undef, <16 x i8> undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 27 for instruction: %V32I8 = call <32 x i8> @llvm.smul.fix.v32i8(<32 x i8> undef, <32 x i8> undef, i32 3)
@@ -214,8 +214,8 @@ define i32 @smul(i32 %arg) {
; BTVER2-NEXT: Cost Model: Found an estimated cost of 114 for instruction: %V16I32 = call <16 x i32> @llvm.smul.fix.v16i32(<16 x i32> undef, <16 x i32> undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I16 = call i16 @llvm.smul.fix.i16(i16 undef, i16 undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call <8 x i16> @llvm.smul.fix.v8i16(<8 x i16> undef, <8 x i16> undef, i32 3)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 53 for instruction: %V16I16 = call <16 x i16> @llvm.smul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 106 for instruction: %V32I16 = call <32 x i16> @llvm.smul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V16I16 = call <16 x i16> @llvm.smul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 90 for instruction: %V32I16 = call <32 x i16> @llvm.smul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I8 = call i8 @llvm.smul.fix.i8(i8 undef, i8 undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V16I8 = call <16 x i8> @llvm.smul.fix.v16i8(<16 x i8> undef, <16 x i8> undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V32I8 = call <32 x i8> @llvm.smul.fix.v32i8(<32 x i8> undef, <32 x i8> undef, i32 3)
@@ -315,8 +315,8 @@ define i32 @umul(i32 %arg) {
; AVX1-NEXT: Cost Model: Found an estimated cost of 114 for instruction: %V16I32 = call <16 x i32> @llvm.umul.fix.v16i32(<16 x i32> undef, <16 x i32> undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I16 = call i16 @llvm.umul.fix.i16(i16 undef, i16 undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call <8 x i16> @llvm.umul.fix.v8i16(<8 x i16> undef, <8 x i16> undef, i32 3)
-; AVX1-NEXT: Cost Model: Found an estimated cost of 53 for instruction: %V16I16 = call <16 x i16> @llvm.umul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
-; AVX1-NEXT: Cost Model: Found an estimated cost of 106 for instruction: %V32I16 = call <32 x i16> @llvm.umul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
+; AVX1-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V16I16 = call <16 x i16> @llvm.umul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
+; AVX1-NEXT: Cost Model: Found an estimated cost of 90 for instruction: %V32I16 = call <32 x i16> @llvm.umul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I8 = call i8 @llvm.umul.fix.i8(i8 undef, i8 undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V16I8 = call <16 x i8> @llvm.umul.fix.v16i8(<16 x i8> undef, <16 x i8> undef, i32 3)
; AVX1-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V32I8 = call <32 x i8> @llvm.umul.fix.v32i8(<32 x i8> undef, <32 x i8> undef, i32 3)
@@ -334,8 +334,8 @@ define i32 @umul(i32 %arg) {
; AVX2-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V16I32 = call <16 x i32> @llvm.umul.fix.v16i32(<16 x i32> undef, <16 x i32> undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I16 = call i16 @llvm.umul.fix.i16(i16 undef, i16 undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V8I16 = call <8 x i16> @llvm.umul.fix.v8i16(<8 x i16> undef, <8 x i16> undef, i32 3)
-; AVX2-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %V16I16 = call <16 x i16> @llvm.umul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
-; AVX2-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V32I16 = call <32 x i16> @llvm.umul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
+; AVX2-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %V16I16 = call <16 x i16> @llvm.umul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
+; AVX2-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %V32I16 = call <32 x i16> @llvm.umul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I8 = call i8 @llvm.umul.fix.i8(i8 undef, i8 undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 14 for instruction: %V16I8 = call <16 x i8> @llvm.umul.fix.v16i8(<16 x i8> undef, <16 x i8> undef, i32 3)
; AVX2-NEXT: Cost Model: Found an estimated cost of 27 for instruction: %V32I8 = call <32 x i8> @llvm.umul.fix.v32i8(<32 x i8> undef, <32 x i8> undef, i32 3)
@@ -448,8 +448,8 @@ define i32 @umul(i32 %arg) {
; BTVER2-NEXT: Cost Model: Found an estimated cost of 114 for instruction: %V16I32 = call <16 x i32> @llvm.umul.fix.v16i32(<16 x i32> undef, <16 x i32> undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I16 = call i16 @llvm.umul.fix.i16(i16 undef, i16 undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call <8 x i16> @llvm.umul.fix.v8i16(<8 x i16> undef, <8 x i16> undef, i32 3)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 53 for instruction: %V16I16 = call <16 x i16> @llvm.umul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 106 for instruction: %V32I16 = call <32 x i16> @llvm.umul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V16I16 = call <16 x i16> @llvm.umul.fix.v16i16(<16 x i16> undef, <16 x i16> undef, i32 3)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 90 for instruction: %V32I16 = call <32 x i16> @llvm.umul.fix.v32i16(<32 x i16> undef, <32 x i16> undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %I8 = call i8 @llvm.umul.fix.i8(i8 undef, i8 undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %V16I8 = call <16 x i8> @llvm.umul.fix.v16i8(<16 x i8> undef, <16 x i8> undef, i32 3)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 45 for instruction: %V32I8 = call <32 x i8> @llvm.umul.fix.v32i8(<32 x i8> undef, <32 x i8> undef, i32 3)
diff --git a/llvm/test/Analysis/CostModel/X86/arith-overflow.ll b/llvm/test/Analysis/CostModel/X86/arith-overflow.ll
index 280a672bb3ac7..b5cdd7d4ecf60 100644
--- a/llvm/test/Analysis/CostModel/X86/arith-overflow.ll
+++ b/llvm/test/Analysis/CostModel/X86/arith-overflow.ll
@@ -1037,8 +1037,8 @@ define i32 @smul(i32 %arg) {
; AVX1-NEXT: Cost Model: Found an estimated cost of 120 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
-; AVX1-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
-; AVX1-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX1-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX1-NEXT: Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
@@ -1056,8 +1056,8 @@ define i32 @smul(i32 %arg) {
; AVX2-NEXT: Cost Model: Found an estimated cost of 62 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
-; AVX2-NEXT: Cost Model: Found an estimated cost of 33 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
-; AVX2-NEXT: Cost Model: Found an estimated cost of 66 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX2-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX2-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 29 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
@@ -1170,8 +1170,8 @@ define i32 @smul(i32 %arg) {
; BTVER2-NEXT: Cost Model: Found an estimated cost of 120 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 56 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
@@ -1275,8 +1275,8 @@ define i32 @umul(i32 %arg) {
; AVX1-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
-; AVX1-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
-; AVX1-NEXT: Cost Model: Found an estimated cost of 104 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX1-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX1-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; AVX1-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
@@ -1294,8 +1294,8 @@ define i32 @umul(i32 %arg) {
; AVX2-NEXT: Cost Model: Found an estimated cost of 60 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
-; AVX2-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
-; AVX2-NEXT: Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX2-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; AVX2-NEXT: Cost Model: Found an estimated cost of 25 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
@@ -1408,8 +1408,8 @@ define i32 @umul(i32 %arg) {
; BTVER2-NEXT: Cost Model: Found an estimated cost of 112 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 52 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 104 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 44 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 88 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 17 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; BTVER2-NEXT: Cost Model: Found an estimated cost of 42 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
diff --git a/llvm/test/Analysis/CostModel/X86/cast.ll b/llvm/test/Analysis/CostModel/X86/cast.ll
index 087fef47db6b9..ae66aaaa2807d 100644
--- a/llvm/test/Analysis/CostModel/X86/cast.ll
+++ b/llvm/test/Analysis/CostModel/X86/cast.ll
@@ -167,8 +167,8 @@ define i32 @zext_sext(<8 x i1> %in) {
; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %F2 = trunc <8 x i32> undef to <8 x i8>
; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %F3 = trunc <4 x i64> undef to <4 x i8>
; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %G = trunc <8 x i64> undef to <8 x i32>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %G1 = trunc <16 x i32> undef to <16 x i16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %G2 = trunc <16 x i32> undef to <16 x i8>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %G1 = trunc <16 x i32> undef to <16 x i16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %G2 = trunc <16 x i32> undef to <16 x i8>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; AVX2-LABEL: 'zext_sext'
@@ -197,8 +197,8 @@ define i32 @zext_sext(<8 x i1> %in) {
; AVX2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F2 = trunc <8 x i32> undef to <8 x i8>
; AVX2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %F3 = trunc <4 x i64> undef to <4 x i8>
; AVX2-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %G = trunc <8 x i64> undef to <8 x i32>
-; AVX2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %G1 = trunc <16 x i32> undef to <16 x i16>
-; AVX2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %G2 = trunc <16 x i32> undef to <16 x i8>
+; AVX2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %G1 = trunc <16 x i32> undef to <16 x i16>
+; AVX2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %G2 = trunc <16 x i32> undef to <16 x i8>
; AVX2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; AVX512F-LABEL: 'zext_sext'
diff --git a/llvm/test/Analysis/CostModel/X86/min-legal-vector-width.ll b/llvm/test/Analysis/CostModel/X86/min-legal-vector-width.ll
index 46293ab0634ee..9b7f37493d628 100644
--- a/llvm/test/Analysis/CostModel/X86/min-legal-vector-width.ll
+++ b/llvm/test/Analysis/CostModel/X86/min-legal-vector-width.ll
@@ -218,8 +218,8 @@ define void @trunc256() "min-legal-vector-width"="256" {
; VEC256-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %A = trunc <8 x i64> undef to <8 x i32>
; VEC256-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %B = trunc <8 x i64> undef to <8 x i16>
; VEC256-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %C = trunc <8 x i64> undef to <8 x i8>
-; VEC256-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %D = trunc <16 x i32> undef to <16 x i16>
-; VEC256-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %E = trunc <16 x i32> undef to <16 x i8>
+; VEC256-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %D = trunc <16 x i32> undef to <16 x i16>
+; VEC256-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %E = trunc <16 x i32> undef to <16 x i8>
; VEC256-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %F = trunc <32 x i16> undef to <32 x i8>
; VEC256-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
diff --git a/llvm/test/Analysis/CostModel/X86/trunc.ll b/llvm/test/Analysis/CostModel/X86/trunc.ll
index b169e9a9ed1b5..a49bd3bb29c0c 100644
--- a/llvm/test/Analysis/CostModel/X86/trunc.ll
+++ b/llvm/test/Analysis/CostModel/X86/trunc.ll
@@ -117,8 +117,8 @@ define i32 @trunc_vXi16() {
; AVX1-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2i32 = trunc <2 x i32> undef to <2 x i16>
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4i32 = trunc <4 x i32> undef to <4 x i16>
; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8i32 = trunc <8 x i32> undef to <8 x i16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i16>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i16>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i16>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; AVX2-LABEL: 'trunc_vXi16'
@@ -132,8 +132,8 @@ define i32 @trunc_vXi16() {
; AVX2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2i32 = trunc <2 x i32> undef to <2 x i16>
; AVX2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4i32 = trunc <4 x i32> undef to <4 x i16>
; AVX2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V8i32 = trunc <8 x i32> undef to <8 x i16>
-; AVX2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i16>
-; AVX2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i16>
+; AVX2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i16>
+; AVX2-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i16>
; AVX2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; AVX512-LABEL: 'trunc_vXi16'
@@ -162,8 +162,8 @@ define i32 @trunc_vXi16() {
; BTVER2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2i32 = trunc <2 x i32> undef to <2 x i16>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4i32 = trunc <4 x i32> undef to <4 x i16>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8i32 = trunc <8 x i32> undef to <8 x i16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i16>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i16>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i16>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
%i64 = trunc i64 undef to i16
@@ -267,9 +267,9 @@ define i32 @trunc_vXi8() {
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2i32 = trunc <2 x i32> undef to <2 x i8>
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4i32 = trunc <4 x i32> undef to <4 x i8>
; AVX1-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8i32 = trunc <8 x i32> undef to <8 x i8>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i8>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i8>
-; AVX1-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64i32 = trunc <64 x i32> undef to <64 x i8>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i8>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i8>
+; AVX1-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V64i32 = trunc <64 x i32> undef to <64 x i8>
; AVX1-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %i16 = trunc i16 undef to i8
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2i16 = trunc <2 x i16> undef to <2 x i8>
; AVX1-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4i16 = trunc <4 x i16> undef to <4 x i8>
@@ -291,9 +291,9 @@ define i32 @trunc_vXi8() {
; AVX2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2i32 = trunc <2 x i32> undef to <2 x i8>
; AVX2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4i32 = trunc <4 x i32> undef to <4 x i8>
; AVX2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8i32 = trunc <8 x i32> undef to <8 x i8>
-; AVX2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i8>
-; AVX2-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i8>
-; AVX2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64i32 = trunc <64 x i32> undef to <64 x i8>
+; AVX2-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i8>
+; AVX2-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i8>
+; AVX2-NEXT: Cost Model: Found an estimated cost of 18 for instruction: %V64i32 = trunc <64 x i32> undef to <64 x i8>
; AVX2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %i16 = trunc i16 undef to i8
; AVX2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V2i16 = trunc <2 x i16> undef to <2 x i8>
; AVX2-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %V4i16 = trunc <4 x i16> undef to <4 x i8>
@@ -363,9 +363,9 @@ define i32 @trunc_vXi8() {
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2i32 = trunc <2 x i32> undef to <2 x i8>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4i32 = trunc <4 x i32> undef to <4 x i8>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %V8i32 = trunc <8 x i32> undef to <8 x i8>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i8>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 15 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i8>
-; BTVER2-NEXT: Cost Model: Found an estimated cost of 30 for instruction: %V64i32 = trunc <64 x i32> undef to <64 x i8>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V16i32 = trunc <16 x i32> undef to <16 x i8>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 13 for instruction: %V32i32 = trunc <32 x i32> undef to <32 x i8>
+; BTVER2-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %V64i32 = trunc <64 x i32> undef to <64 x i8>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %i16 = trunc i16 undef to i8
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2i16 = trunc <2 x i16> undef to <2 x i8>
; BTVER2-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4i16 = trunc <4 x i16> undef to <4 x i8>
More information about the llvm-commits
mailing list